T lymphocyte ("T cell") antigen receptors ("TCR") recognize endogenously processed fragments of antigens that are presented to T cells in association with major histocompatibility complex ("MHC") class I or class II molecules.
An individual's T cells recognize and are activated by protein antigens only if a fragment of the antigen is properly presented on the surface of a target cell. The antigen presentation process that allows an antigen to be recognized by a T cell requires that the antigen be associated with either (MHC) class I histocompatibility molecules for presentation to cytotoxic T lymphocytes ("CTLs") or class II histocompatibility molecules for presentation to helper T cells. Other T cell subsets such as .gamma./.delta. (gamma-delta) T cells (CD4.sup.-, CD8.sup.-) may recognize alternate "peptide presenting" molecules not encoded in the MHC, such as CD1, etc. The subset of T cells denoted CD8.sup.+ recognize antigenic determinants/epitopes that are associated with class I histocompatability molecules. The other subset of T cells, CD4.sup.+ cells, recognize antigenic determinants/epitopes that are associated with class II histocompatibility molecules. The antigenic determinants/epitopes that are presented on the surface of cells in association with MHC molecules are also known as T cell epitopes.
The study of CD8.sup.+ T cell recognition of target cells has been extensive since the early 1970's when Zinkernagel and Doherty demonstrated that CTL recognition of viral-infected autologous target cells requires the presence of self class I MHC molecules. Thus such recognition of target cells by CD8.sup.+ T cells is referred to as being MHC class I-restricted. Zinkernagel, R. M., et al., Adv. Immunol. 27:51 (1979); Doherty, P. C., et al., Adv. Cancer Res. 42:1 (1984); and Zinkernagel, R. M., et al., Nature 248:701 (1974), the disclosures of which are incorporated herein by reference. It was later shown that virus-specificity of CTL's is directed against viral protein-derived peptide sequences that are presented by infected cell MHC class I molecules to CD8.sup.+ T cells. See, for example, Townsend, A., et al., Cell 42:457 (1985) and Townsend, A., et al., Cell 44:959 (1986), the disclosures of which are incorporated herein by reference.
As noted above, it is not the entire antigen that is presented by target cells and recognized by CD8.sup.+ cells, but rather what is presented and recognized are small endogenously processed peptides that are generated from antigens by intracellular degradation pathways in either the cytosol or the endoplasmic reticulum ("ER") of the target cell. Such processed peptides bind to newly synthesized class I heavy chain-.beta..sub.2 -microglobulin heterodimers in the ER. See. for example, Yewdell, J. W., et al., Science 244:1072 (1989); Townsend, A., et al., Cell 62:285 (1990); and Nuchtern, J. G., et al., Nature 339:223 (1989), the disclosures of which are incorporated herein by reference. The processed peptide is bound to the class I heavy chain-light chain dimer molecule via the class I antigen binding site/peptide cleft. The complex thereby generated is a transport competent trimer as reported by Yewdell, J. W., et al., Science 244:1072 (1989); Townsend, A., et al., Cell 62:285 (1990); and Nuchtern, J. G., et al., Nature 339:223 (1989). This class I histocompatibility molecule-processed peptide complex is then expressed on the surface of the target cell where it may be ultimately recognized by T cell clonotypic receptors on CD8.sup.+ cells in conjunction with CD8 accessory molecules. See, Rotzschke, O., et al., Nature 348:252 (1990); Van Bleek, G. M., et al., Nature 348:213 (1990); Rotzschke, O., et al., Science 249:283 (1990); and Falk, K., et al., Nature 348:248 (1990), the disclosures of which are incorporated herein by reference.
Recently, peptides have been isolated from the antigen binding sites of human and murine class I and class II molecules and directly sequenced. Two principal methods have been used to isolate such peptides. In one of the two methods total cellular extraction of such peptides is carried out in pH 2.0 trifluoroacetic acid ("TFA"). This method results in cell cytolysis and release of total cytosolic peptides, only a fraction of which are actually class I-related. This method also typically employs protease inhibitors since cell cytolysis results in the release of proteolytic enzymes that can alter or destroy peptides of potential interest. See, Rotzschke, O., et al., Nature 348:252 (1990), and Falk, K., et al., Nature 348:248 (1990), the disclosures of which are incorporated herein by reference. The second isolation method entails acid denaturation of immunoaffinity purified class I-peptide complexes. By contrast with the first method, the second method of peptide isolation is highly class I selective, and even class I allele specific since monoclonal antibodies directed against individual class I allotypes can be used to immunopurify class I complexes. By this latter approach, the majority of known class I-bound peptide sequence data has been acquired. See, for example, Van Bleek, G. M., et al., Nature 348:213 (1990); Rotzschke, O., et al., Science 249:283 (1990); Madden, D. R., et al., Nature 353:326 (1991); Jardetzky, T. S., et al., Nature 351:290 (1991); and Nikolic-Zugic, J., et al., Immunol Rev. 10:54 (1991), the disclosures of which are incorporated herein by reference.
The main drawback of these two methods is that since both require cell cytolysis, a large number of starting cells (10.sup.9 -10.sup.11) are required from which peptides are extracted after cellular cytolysis in order to obtain sequence grade quantities (approximately 1 pM) of specific peptide. Therefore the application of such techniques are limited to cell types which readily adapt to in vitro cell culture and which proliferate sufficiently well to allow such high cellular yields.
Methods of isolating class I peptide complexes are additionally relevant because CD8.sup.+ lymphocytes have emerged as being potentially useful in the development of anti-tumor vaccines, which vaccines will ideally provoke anti-tumor immune responses in individuals. To that end, tumor infiltrating lymphocytes (TILs) have been found to be important agents in the generation of cellular immunity through their identification in spontaneously regressing lesions in some patients as reported by Kornstein, M. J., et al. Cancer Res. 43:2749 (1983), the disclosure of which is incorporated herein by reference. TILs are also frequently found in non-regressing lesions and when present in high numbers are correlated with a better clinical prognosis. Van Duinen, S. G., et al., Cancer Res. 48:1019 (1988), the disclosure of which is incorporated herein by reference. Numerous studies have shown that such TILs display potent anti-melanoma cytolytic activity when they are cultured in vitro with interleukin-2. See, for example, Lotze, M. T., Pigment Cell 10:163 (1990), and Rosenberg, S. A., et al., N. Eng. J. Med. 319:1676 (1988). Anti-melanoma cytolytic activity is typically associated with CD8.sup.+ TIL subpopulations which recognize tumor cells in a class I--restricted manner. The HLA class I antigen, HLA-A2, appears to represent the most common class I restriction element for human melanoma TIL, however, other HLA class I antigens such as HLA-A1, -A10, -A24, -A31, -B44, -B50, and -CW7 have also been identified. The identification of such restriction elements may be important in the development of effective melanoma vaccines.
During the last several years, a number of further studies have been conducted on the autologous CD8.sup.+ T cell-mediated response to human melanoma. See, for example, Parmiani, G., et al., J. Natl. Cancer Inst. 82:361 (1990) and Van den Eynde, B., et al., Int. J. Cancer 44:634 (1984), the disclosures of which are incorporated herein by reference. The emerging picture indicates that melanomas express multiple T cell epitopes, some of which are unique to a given tumor, while others are shared by allogeneic, HLA-matched melanomas. See, for example, Anichini, A., et al., J. Immunol 142:3692 (1989); Wolfel, T., et al., Eur. J. Immunol. 24:759 (1994); and Crowley, N. J., et al., J. Immunol. 146:1692 (1991), the disclosures of which are incorporated herein by reference. These epitopes appear to represent short 9-10 amino acid peptides derived from tumor-associated antigens that are presented by MHC class I antigens to CD8.sup.+ T cells. See, for example, Traversari, C., et al., J. Exp. Med. 176:1453 (1991) and Kawakami, Y., et al., J. Exp. Med. 180:347 (1994), the disclosures of which are incorporated herein by reference. While many class I alleles have been reported to represent restriction elements for tumor-reactive CD8.sup.+ T cells, as reported by Hom, S. S., et al., J. Immunother. 10:153 (1991), the disclosure of which is incorporated herein by reference, the HLA-A2.1 allele, which is expressed by 45% of melanoma patients, appears to play an immunodominant role in presenting melanoma epitopes as reported by Crowley, N. J., et al., J. Immunol. 146:1692 (1991). As will be shown herein, at least six different CD8.sup.+ T cell-defined epitopes appear to be expressed by multiple HLA-A2.sup.+ melanomas. The identification and sequencing of these individual epitopes should allow for the design and testing of peptide-based immunotherapies for the treatment of melanoma.
Presentation of MHC class I-restricted tumor antigens to CD8.sup.+ T lymphocytes was classically assumed to be presented by the cell naturally expressing the antigens, such as the tumor cell itself. However, recent studies suggest that tumor cells are poor antigen presenting cells, and "professional" antigen presenting cells may be essential for induction of not only CD4.sup.+ T cells but also for a CD8.sup.+ T cell-mediated antitumor response. Huang, A. Y. C., et al., Science (Wash. DC) 264:961 (1994), the disclosure of which is incorporated herein by reference, reported that the in vivo priming of MHC class I-restricted responses involves a transfer of tumor antigens to a host bone marrow-derived dendritic cell ("IBM-DC") and subsequent presentation to CD8.sup.+ T cell effectors. A correlation between the number of presumptive dedritic cells ("DC") infiltrating a cancer and longer patient survival or reduced frequency of metastatic disease have been observed for a variety of cancer types. See, Becker, Y., Vivo 7:187 (1993), the disclosure of which is incorporated herein by reference. DC were first described by Steinman, R. M., et al., J. Exp. Med. 137:1142 (1973), the disclosure of which is incorporated herein by reference, and are known to be highly specialized antigen-presenting cells and to be the principal activators of resting, naive T cells in vitro and in vivo. DC are further capable of efficiently transporting antigens from the periphery to lymphoid organs. See, Levin, D., et. al., J. Immunol 151:6742 (1993); Constant, S., et al., J. Immunol 154:4915 (1995); and Cohen, P. J., et al., Eur. J. Immunol 24:315 (1994), the disclosures of which are incorporated herein by reference. DC pulsed with ovalbumin or viral peptides, lymphoma-derived surface Ig, or tumor-derived synthetic epitopes and then injected as a vaccine have been shown to immunize naive animals against a subsequent viral or tumor challenge. See, Porgador, A., et al., J. Exp. Med. 182:255 (1995); Takahashi, H. Y., et al., Int. Immunol. 5:849 (1993); Flamand, V., et al., Eur. J. Immunol 24:605 (1994); Mayordomo, J. I., et al., Nature Med. 1:1297 (1995); and Celuzzi, C. M., et al., J. Exp. Med. 183:283 (1996), the disclosures of which are incorporated herein by reference.
Development of antitumor vaccines based on MHC-presented tumor peptide epitopes capable of being recognized by T cells is limited due to the fact that defined tumor-derived T cell epitopes are currently available for study in only a limited number of human tumor types (i.e., melanoma, ovarian, and breast carcinoma) as reported by Storkus, W. J., et al., Biologic Therapy of Cancer 2nd ed. DeVita, V. T., et al., editors. J. B. Lippincott Co., Philadelphia, Pa. 66-77 (1995), the disclosure of which is incorporated herein by reference. Additionally, the vast majority of the histologic subtypes of cancers express as yet undefined shared tumor epitopes and individual patients may also express idiotypic determinants unique to that patient.
It is also difficult to extend the range of the search for biologically relevant allo-, viral-, and tumor-specific T cell epitopes to cell types that adapt poorly to tissue culture or which proliferate slowly in vitro. Accordingly there is a need for methods that will remove T cell epitopes from a greater range of cell types. In doing so, the development of peptide-based immunotherapies for the treatment of patients with melanoma and other diseases may be furthered. In particular, there is a need for methods to develop protocols to treat cancer patients whose tumors express as yet uncharacterized T cell epitopes.